Groq Pricing Calculator
Updated July 2026Pure-usage per-token serverless + per-hour transcription + per-1M-character text-to-speech + per-use tools + sales-led enterprise
Groq pricing: Groq pricing 2026: Llama 3.1 8B at $0.05/$0.08, GPT OSS 120B $0.15/$0.60, Llama 3.3 70B $0.59/$0.79, Whisper Turbo $0.04/hr, 50% Batch + cached, LPU latency. Use the free calculator below to enter your usage and get an instant all-in monthly estimate — including overages — and see which plan is cheapest for your needs.
How Groq prices: Pure-usage per-token serverless + per-hour transcription + per-1M-character text-to-speech + per-use tools + sales-led enterprise.
How Groq prices
Pure-usage per-token serverless + per-hour transcription + per-1M-character text-to-speech + per-use tools + sales-led enterprise
Best fit — auto-selected: Free tier
$14.80/mo— the cheapest plan that fits; adjust usage and it re-picks
Your usage
💡 1M input tokens ≈ 750k words of prompt/context; a busy chat or RAG app runs 100–500M/mo
Groq serverless input price varies by model. Cached input tokens bill at 50% of these rates; the Batch API applies a separate 50% discount.
💡 Output is the model's generated response; typically 20–40% of input volume for chat/RAG
Per-model output rate from the Groq rate card. Match this to the same model you picked for the input rate.
💡 Whisper bills per hour of audio transcribed (10-second minimum per request); a call center runs hundreds–thousands of hours/mo
Whisper Large v3 is higher-accuracy and slower; Turbo is 2.8× cheaper and faster with slightly lower accuracy.
💡 Built-in web search is billed per 1,000 requests; an agent loop can fire thousands–tens of thousands per month
Basic search is $5 per 1,000 requests ($0.005 each); Advanced search is $8 per 1,000 ($0.008 each). Website visits are $1 per 1,000.
💡 Code execution (code interpreter) is billed at $0.18 per hour of compute
Compare Plans at Your Usage
Monthly Estimate
Free tier
$14.80
per month
Cost Breakdown
Annual
$177.60
Why Free tier — $14.80/mo
- Free tier at $14.80/mo.
- How we got there: 200 M tokens/mo (Input tokens) × 0 $/1M in (Model (input rate)) = 10 $.
- How we got there: 60 M tokens/mo (Output tokens) × 0 $/1M out (Model (output rate)) = 5 $.
- How we got there: 0 hours/mo (Audio transcription) × 0 $/hr (Transcription model (per hour)) = 0 $.
- How we got there: 0 requests/mo (Web search requests) × 0 $/req (Search depth (per request)) = 0 $.
- Input tokens (10 $ × $1) is 68% of your bill — the lever that matters most.
Budget range
Plan for usage swings, not just today's estimate.
Conservative
$9.80
100 M tokens/mo/mo
Expected
$14.80
200 M tokens/mo/mo
Aggressive
$24.80
400 M tokens/mo/mo
Need a calculator like this on your pricing page?
Embed interactive pricing calculators on your website to help customers understand costs and boost conversions.
Get StartedGroq plans at a glance
| Plan | Price | Best for |
|---|---|---|
| Free tier | Free | Developers, evaluation, low-volume prototypes — rate-capped, no card required |
| Developer (pay-as-you-go) | Usage-based | Latency-sensitive production AI apps — full model catalog, credit-card billing |
Prices shown are list rates as of July 2026. Enter your usage above for an all-in estimate including overages and add-ons.
About this Groq calculator
This calculator estimates your Groq cost from publicly available pricing. Actual costs may vary with your specific agreement, volume discounts, and usage patterns — always verify on the provider's official pricing page for the most current rates.
Groq pricing — frequently asked questions
How much does Groq cost per month? ▼
Groq has no monthly subscription fee — you pay only for the tokens you consume on serverless inference, plus per-hour audio transcription, per-use built-in tools, and per-session agentic infrastructure. A small chat application using Llama 3.1 8B at 30M input + 10M output tokens would cost roughly $2.30/month — Groq's per-token rates are among the lowest in the market for small-model inference.
What are Groq's per-token rates for popular models? ▼
Per 1M tokens (input/output): Llama 3.1 8B Instant $0.05/$0.08 at 840 TPS; GPT OSS 20B $0.075/$0.30 at 1,000 TPS; Llama 4 Scout $0.11/$0.34 at 594 TPS; Qwen3 32B $0.29/$0.59 at 662 TPS; GPT OSS 120B $0.15/$0.60 at 500 TPS; Llama 3.3 70B Versatile $0.59/$0.79 at 394 TPS. Throughput numbers are published alongside prices to make the latency-cost trade-off transparent.
Does Groq have a free tier? ▼
Yes — Groq's free tier offers rate-capped access to most models for experimentation. Throughput limits apply (low requests-per-minute and tokens-per-minute caps), with sustained production usage requiring a paid plan. Free tier credit-card-on-file is not required for evaluation.
How does Groq's LPU differ from GPU-based inference? ▼
Groq's LPU (Language Processing Unit) is bespoke silicon designed exclusively for inference. Unlike GPUs which use HBM (high-bandwidth memory) and branch prediction, the LPU uses on-die SRAM and deterministic execution. The trade-off is lower per-chip memory capacity but vastly higher single-stream throughput — enabling 800+ tokens/second on Llama 3 8B and 500+ TPS on 120B-class models.
Is this Groq pricing calculator free and accurate? ▼
Yes — it's completely free with no signup. It uses the latest publicly available Groq pricing (verified July 2026). Actual costs may vary with volume discounts or enterprise terms — always confirm on Groq's official pricing page.